Cephalgo and the EIC Accelerator: Tracking Emotions in Depression

Cephalgo, registered in Strasbourg in 2021, develops AI to track the emotional state of people with major depressive disorder between psychiatric appointments, so doctors can see whether treatment is working. It was selected in the EIC Accelerator's March 2023 cut-off and received an EU grant of €2.50 million for ADAPTE, "A novel and accurate emotion recognition system for real-time and continuous patient monitoring in psychiatry", a €3.57 million project that ran from November 2023 to April 2026, according to CORDIS. The project's objective says Cephalgo "has solved this challenge" with speech analysis and an AI emotion recognition algorithm that also "predicts the best treatment".

The project has ended, but CORDIS holds only its first periodic report, to April 2025. It lists enhanced speech-based emotion recognition, a tested "patient conversational pathway v1.0", engagement modules validated with clinicians, and partnerships with psychiatric hospitals "covering 6,000 MDD patients", according to the CORDIS reporting page. Its next steps include larger datasets and "ensure product-market fit". No clinical accuracy data, launch or final report has been published.

Where Cephalgo Stands Now

Cephalgo is active in the French company register, which puts its workforce at three to five employees, according to recherche-entreprises. Its accounts are filed but not published in the free data. The English page of its website returned "Page Not Found" when checked in September 2026. On the partner page of the Horizon Europe OPADE project, Cephalgo still describes itself as developing "the first patient monitoring device in psychiatry using EEG and AI", while its EIC project relies on speech analysis on smartphones.

The Technology: Listening for Mood

The project objective says a psychiatrist may see a patient only every three to four months and has no way to know between sessions whether treatment is working or whether relapse or suicide risk is rising. Cephalgo's software analyses patients' natural speech with time-series AI to estimate emotional state, and adds engagement tools to keep patients on treatment. The report cites dropout rates above 60% and aims to support adherence of 75%.

What the EIC Award Promised, and What the Record Shows

The award promised an accurate emotion recognition system for continuous monitoring in psychiatry. At the halfway point there was a first version of the patient pathway and hospital partnerships. The claim that the system "detects emotions with high accuracy" is not backed by any published figure, and what happened in the project's final year is not public.

1. Speech emotion AIPartialEnhanced, per first report.
2. Patient appPartialPathway v1.0 tested.
3. Clinical validationUnverifiedNo accuracy published.
4. Final reportingMissingOnly period 1 public.

The Money

The EU paid 70% of the EIC project and a further €681,375 through OPADE, according to CORDIS participant data. Cephalgo's revenue, losses and private investors are not public in free sources.

What Cannot Be Checked

Why Europe Should Care

The project objective cites 150 million people with mental disorders in Europe and too few psychiatrists. Tools that let clinicians see between visits could help, if they work. The public has been told the problem is solved but has not been shown the evidence.

The Verdict

Cephalgo and the EIC have not been asked for comment for this article. It is based entirely on public records and published statements, linked throughout.